AI photos of each garment on several Indian body types, for small fashion brands
TL;DR Drapeview lets a small fashion label photograph a garment once and then show it, labelled as AI-made, on five or six Indian body types from petite to plus size. Shoppers pick the body nearest their own. Brands pay a monthly fee per catalogue, betting that fewer orders come back for fit.
Clothes bought online come back far more often than most goods. Academic datasets put fashion return rates anywhere from 13% to 96%, averaging around 53%[1]. I found no public figure for India, which tells you something in itself.
In a 2023 survey by a big search company and a polling firm of US shoppers, 42% said they didn't feel represented by the models in product images, and 59% had been unhappy with clothing that looked different on them than expected[2].
Small Indian labels feel this the most. They usually shoot one slim model in one size, sell S to 3XL and eat the reverse shipping when a kurta that looked loose on the page turns out tight at the hips.
Drapeview is a catalogue tool and it doesn't try to be a mirror. A brand uploads its usual shoot plus the garment's measurement chart, and Drapeview renders that piece on a fixed set of reference bodies across Indian heights, builds and skin tones. Those bodies come from consented, paid model shoots. Each image is tagged "AI-generated, size L on a 5'3" body" so nobody mistakes it for a photograph. Shoppers tap the body closest to theirs on the product page, and the brand gets a monthly report linking which images were viewed to which orders came back.
The seed pitched ultra-realistic try-on from each shopper's own photo, and I reframed it. That route needs body photos from strangers, consent under the DPDP Act[3] and render quality good enough to trust, while the biggest platforms already give a version away free. Studies of try-on users also show intrusiveness puts people off more than general privacy worries do[4]. A fixed set of bodies sidesteps all of that and still answers the shopper's real question - what will this look like on someone built like me?
The shopper uses it free. The payer is a D2C apparel brand with roughly 50–500 live styles, selling on its own hosted web store and on marketplaces. Brand founders already pay for model shoots and for returns logistics, so the pitch is a monthly subscription sized by styles rendered and priced below a single extra shoot day. I found no public price list for Indian AI model-photo tools, so treat the price as an assumption to test.
A big search company showed in June 2023 that one garment image could be rendered on real models from XXS to 4XL[2]. That kind of generation is now available as cloud models a small team can tune. And India has roughly 290–300 million online shoppers, close to 45% of them women[5], with tonnes of first-time buyers in smaller towns.
There's no public figure for what Indian brands spend on catalogue imagery or fit tools. The adjacent signal is IBEF's projection of trend-led fashion at ₹68,456–85,570 Cr by 2028, with more than half of it sold online[5].
IBEF doesn't name its underlying sources though. Use it as a rough guide to how many brands need product images and not as your market.
One search company's try-on renders apparel on diverse models inside its shopping results, launched for US shoppers[2]. One lock-screen content company runs an AI shopping feed and app in India[6], but it sells attention and not catalogue imagery for brands. Large marketplaces build their own tools for their own listings.
What's missing is a low-cost service a 20-person label can use on its own store and on marketplace listings alike.
| Difficulty | Moderate (36/100) |
| Starting capital | ₹60 L–1.2 Cr |
| Time to launch | 6–9 months |
Generation is the easy bit and accuracy is the hard one. Saree pleats, dupatta drape and stretch fabrics are exactly where generated images go wrong, and an image that flatters a garment will push returns up instead of down. You need a measurement pipeline that follows each brand's size chart and a human check before any image goes live.
No sector licence applies to selling image software to brands. Consent and honest images matter more here.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
AI is the product here, so the question is where it helps and where it needs a person.
Don't trust renders of saree pleats, dupatta drape or stretch fabric without a human check. A flattering image that misleads on fit pushes returns up, not down.
Want to work out the AI layer for Drapeview? Talk to Yash →
I'd back this for a founder who knows apparel production and size grading as well as image models. A pure AI team will underrate how often garments differ from their own size charts. Before you build much, prove one thing with five brands - fit-related returns fall on styles with Drapeview images compared with matched styles without them.
Plan for roughly ₹60 L–1.2 Cr, mostly on consented reference shoots, the rendering pipeline and human checks.
No sector licence. You need written consent from reference models under the DPDP Act and clearly labelled, accurate images under consumer law.
Yes, small fashion brands everywhere shoot one model in one size. Use reference bodies that match local shoppers, and check local rules on labelling AI-made images.
The free Drapeview validation kit has the line-by-line capital breakdown, the effort scoring, the consent and advertising checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for brand founders, and 5 assumptions to test. Download the kit →
Thinking about building Drapeview or something close to it? The make-or-break work is proving to brands, with their own return data, that the images change what comes back. I've helped founders set up exactly this kind of before-and-after test. If you need a hand, you know where to find me. Talk to Yash →
For illustration only. Names are fictional and ideas hypothetical; sourced facts are real but may be out of date. Nothing here is legal, financial or professional advice, and Yash Khatri is not personally liable for any use of it. Read the full disclaimer.